根据之前提到的需求和注意事项，以下是满足要求的一个完整的Python代码示例。此代码仅为演示目的，在实际应用中请务必遵守社交媒体平台的服务条款并尊重用户隐私保护法规。 ### 完整 Python 代码 示例 #### 第一步：安装依赖包 首先确保已经安装了必要的Python库： ```bash pip install selenium pandas requests beautifulsoup4 ``` #### 第二步：编写主程序 ##### 导入模块 ```python import time from datetime import datetime, timedelta from selenium import webdriver from selenium.webdriver.common.by import By from selenium.webdriver.chrome.service import Service as ChromeService from webdriver_manager.chrome import ChromeDriverManager from bs4 import BeautifulSoup import pandas as pd ``` ##### 初始化WebDriver ```python def init_driver(): options = webdriver.ChromeOptions() # 添加无头模式和其他选项可以根据需求调整 options.add_argument("--disable-blink-features=AutomationControlled") driver = webdriver.Chrome(service=ChromeService(ChromeDriverManager().install()), options=options) return driver ``` ##### 登录函数（需手动输入账号密码） ```python def login_weibo(driver): url_login = 'https://passport.weibo.cn/signin/login' driver.get(url_login) username_input = driver.find_element(By.NAME,'username') password_input = driver.find_element(By.NAME,'password') USERNAME = "your_username" # 替换为您的微博用户名 PASSWORD = "your_password" # 替换为您的微博密码 username_input.send_keys(USERNAME) password_input.send_keys(PASSWORD) submit_button = driver.find_element(By.XPATH,"//div[@class='info_list']/div[6]/a") submit_button.click() print("正在尝试自动登录...") time.sleep(5) # 等待几秒让页面加载完毕 ``` ##### 获取特定时间段内的帖子列表 ```python def fetch_posts_by_keyword_and_time(keyword, start_date, end_date): base_url = f"https://s.weibo.com/weibo?q={keyword}&typeall=1&suball=1" date_format_str = "%Y-%m-%d" posts_data = [] while True: current_page_url = build_search_query(base_url, keyword, start_date.strftime(date_format_str), end_date.strftime(date_format_str)) soup = load_page(current_page_url) if not check_if_more_pages(soup): break extract_post_details(posts_data, soup) next_day = start_date + timedelta(days=1) if next_day > end_date: break else: start_date = next_day df = pd.DataFrame(posts_data) save_to_file(df, filename="weibo_earthquake.csv") def build_search_query(base_url, keyword, from_date=None, to_date=None): query_parts = [base_url] if from_date and to_date: query_parts.append(f"&timescope=custom:{from_date}:{to_date}") return ''.join(query_parts) def load_page(page_url): driver.get(page_url) html_content = driver.page_source return BeautifulSoup(html_content, features="lxml") def check_if_more_pages(soup_object): pagination_links = soup_object.select('.next-page a') or [] return len(pagination_links)>0 def extract_post_details(data_container, soup): post_elements = soup.select('div[action-type="feed_list_item"]') for element in post_elements: try: post_info = { "id": get_attribute(element, ".WB_from", attr_name="data-id"), "author_id": get_attribute(element, ".name", href=True), "nickname": get_text_or_none(element, '.name'), "post_time": clean_datetime(get_text_or_none(element, ".WB_from")), "content": get_text_or_none(element, ".txt"), "reposts_count": int(extract_number_from_string(element.select_one(".repeat").text)), "comments_count": int(extract_number_from_string(element.select_one(".comment").text)), "likes_count": int(extract_number_from_string(element.select_one(".like").text)), "user_nickname": get_user_profile_field(element, field=".profile .name"), "verified_type": get_user_profile_field(element, field=".profile span"), "location": get_user_profile_field(element, field=".profile em.location"), "followers_num": convert_to_int(get_user_profile_field(element, field=".profile .follow strong")), "following_num": convert_to_int(get_user_profile_field(element, field=".profile .fans strong")), "tweets_num": convert_to_int(get_user_profile_field(element, field=".profile .tweet strong")), "account_created_at": parse_account_creation_date(get_user_profile_field(element, field=".profile")) } data_container.append(post_info) except Exception as e: print(e) def get_user_profile_field(post_elem, field='.'): profile_section = post_elem.find_next_sibling(class_='card-user') value = None if not profile_section else profile_section.select(field)[0].get_text(strip=True) if profile_section.select(field) else '' return '' if isinstance(value, list) else str(value).strip() def get_attribute(elem, selector, attr_name=None, href=False): selected_elem = elem.select(selector) result = "" if selected_elem: target_attr = ("href" if href else "") or (attr_name or "") result = selected_elem[-1][target_attr] if target_attr else selected_elem.text.strip() return result def get_text_or_none(elem, css_selector): elements = elem.select(css_selector) return "" if not elements else elements[0].get_text(separator='\n').strip() def extract_number_from_string(text_with_numbers): numbers_only = ''.join([ch for ch in text_with_numbers if ch.isdigit()]) return max(int(numbers_only), default=0) def clean_datetime(dt_string): dt_cleaned = re.sub(r'\D', '', dt_string)[:len("YYYY-MM-DD HH:mm:ss")] try: parsed_dt = datetime.strptime(dt_cleaned,"%Y%m%d%H%M%S") return parsed_dt.isoformat() except ValueError: return "" def convert_to_int(mixed_value): num_as_str = mixed_value.replace(',','').replace('+','').replace('万','e+4') return round(float(num_as_str)) def parse_account_creation_date(profile_field_text): patterns=[ r"(?P<year>\d{4})年(?P<month>\d{2})月", ] matches=[match.groupdict()for pattern in patterns for match in map(re.compile(pattern).search,[profile_field_text])] dates_found=[datetime(**group)for group in matches] oldest_registration=min(dates_found,default=datetime.now()) registration_iso8601=oldest_registration.date().isoformat() return registration_iso8601 def main(): global driver driver = init_driver() login_weibo(driver) start_d,end_d=(lambda s,e:[datetime(year=int(y), month=int(m), day=int(d)).date()\ for y,m,d in zip(*[[int(i)for i in d.split('-')] \ for d in[s,e]])])(*["2025-01-07","2025-01-17"]) fetch_posts_by_keyword_and_time("#微博定日6.8级地震#",start_d,end_d) if __name__ == "__main__": main() def save_to_file(df,filename): df.to_csv(filename,index=False,encoding='utf-8-sig') print("已完成数据采集并保存到CSV文件.") driver.quit() ``` 该段代码实现了对给定日期范围内有关“#微博定日6.8级地震#”主题下的所有公开可见的微博文章的基本信息检索，并将其导出至名为`weibo_earthquake.csv` 的 CSV 文件之中。 注意：由于新浪微博可能会不断更新其前端界面设计和技术架构，因此上述CSS选择器路径等细节在未来版本中可能不再适用，请及时跟进官方文档或社区讨论来维护和完善代码逻辑。此外，出于安全性和合法性考量，建议仅用于学习研究用途而非商业牟利。